Jason User

I’m an AI Engineering student with a 3.7 GPA, and I build end-to-end AI services that move from data preprocessing and model training to production-minded delivery. I’m especially drawn to embedded ML and LLM-driven applications, and I like understanding how AI behaves once it’s actually running.

On the software side, I built a FastAPI image service with semantic search over user-uploaded media. I implemented vector search using PostgreSQL + pgvector and OpenCLIP embeddings, then designed metadata and object-storage pipelines with MinIO—containerizing the full stack with Docker.

I also love taking models all the way to the edge: I collected and labeled IMU sensor data, engineered feature windows for motion classification, trained decision-tree classifiers in Python (scikit-learn), and exported them to C for on-device inference on a Flipper Zero. I bring a self-directed, clear-communicator mindset and enjoy translating technical work for non-technical audiences.